01 を動かす〜音声で対話できる自動化されたアシスタント / Running 01 - Automated Assistant with Voice Interaction

早稲田大学大学院経営管理研究科「プロンプトエンジニアリング ─ 生成AI の応用」2025 春のオンデマンド教材 第13回で使用したスライドです。

Transcript

1. Generated by Stable Diffusion 3.5 — AI 2025 13 01 ∼ (WBS) 2025 13 01 ∼ — 2025-05 – p.1/21
2. https://speakerdeck.com/ks91/collections/prompt-engineering-2025-spring 2025 13 01 ∼ — 2025-05 – p.2/21
3. ( 20 ) 1 • 2 • 3 • 4 • 5 • 6 RPG • 7 “September 12th” • 8 • 9 • 10 ∼ • 11 Linux (Windows )(Mac ) • 12 Open Interpreter ∼ • 13 01 ∼ • 14 AGI (Artificial General Intelligence) 7 (4/28 ) / (2 ) OK / 2025 13 01 ∼ — 2025-05 – p.3/21
4. 01 Raspberry Pi OS (64-bit) on Raspberry Pi 4B (ARM Linux) ( ) macOS Sequoia 15.4 15.5 on Parallels(macOS) (ARM Darwin) ARM Ubuntu 24.04 on WSL(ARM Windows 11) on Parallels(macOS) (ARM Linux ) ARM Ubuntu 24.04 on Lima(macOS) (ARM Linux ) x86 Ubuntu 24.04 on Lima(macOS) (Intel Linux segmentation fault ) macOS (Linux Darwin) 2025 13 01 ∼ — 2025-05 – p.4/21
5. macOS macOS zsh . . . bash % chsh -s /bin/bash “.bash_profile” ( ) .bashrc ( ) source ~/.bashrc Homebrew zsh .bash_profile eval "$(/opt/homebrew/bin/brew shellenv)" source ~/.bashrc 2025 13 01 ∼ — 2025-05 – p.5/21
6. API Key .bashrc .bashrc export POETRY_VERSION=1.8.5 2025 13 01 ∼ — 2025-05 – p.6/21
7. 01 Open Interpreter (CLI ) ( macOS ) : Intel Linux 2025 13 01 ∼ — 2025-05 – p.7/21
8. 01 (0) Python3.11 (1) livekit (2) poetry (3) Rust compiler (4) (5) 01 2025 13 01 ∼ — 2025-05 – p.8/21
9. (0) Python3.11 01 Python 3.10 3.11 Python 3.11 ( ) Ubuntu $ sudo apt install -y software-properties-common $ sudo add-apt-repository ppa:deadsnakes/ppa $ sudo apt update $ sudo apt install python3.11 python3.11-venv python3.11-dev macOS $ brew install python@3.11 2025 13 01 ∼ — 2025-05 – p.9/21
10. (0) Python3.11 Python3 $ python3.11 -m venv 01env $ source 01env/bin/activate ( python3 python3.11 ) (01env) $ $ source 01env/bin/activate ↑ 2025 13 01 ∼ — 2025-05 – p.10/21
11. (1) livekit AI ( ) https://01.openinterpreter.com/server/livekit “livekit-server is installed” 2025 13 01 ∼ — 2025-05 – p.11/21
12. (2) poetry Python https://python-poetry.org/docs/#installing-with-the-official-installer ( Linux, macOS, WSL) “Poetry (1.8.5) is installed now. Great!” .bashrc export PATH="/home/ /.local/bin:$PATH" $ exec $SHELL -l $ poetry --version 2025 13 01 ∼ — 2025-05 – p.12/21
13. (3) Rust compiler Rust compiler $ curl https://sh.rustup.rs -sSf | sh / “Rust is installed now. Great!” $ exec $SHELL -l 2025 13 01 ∼ — 2025-05 – p.13/21
14. (4) GitHub 01 ( ) $ git clone https://github.com/OpenInterpreter/01.git ( ) $ git clone https://github.com/ks91/01.git 2025 13 01 ∼ — 2025-05 – p.14/21
15. (5) 01 Ubuntu $ sudo apt install portaudio19-dev macOS $ brew install portaudio $ cd 01/software $ poetry install $ poetry install “No dependencies to install or update” 2025 13 01 ∼ — 2025-05 – p.15/21
16. 01 01 Light 01 01 Light 2025 13 01 ∼ — 2025-05 – p.16/21
17. ( ) 01/software $ cd source/server/livekit $ nano multimodal.py Python model = openai.realtime.RealtimeModel( instructions=instructions, voice="shimmer", temperature=0.6, model="gpt-4o-realtime-preview", modalities=["audio", "text"], api_key=openai_api_key, base_url="wss://api.openai.com/v1", ) 2025 13 01 ∼ — 2025-05 – p.17/21
18. 01 Light (Wi-Fi ) iOS / iPadOS https://apps.apple.com/jp/app/01-light/id6601937732 android https://play.google.com/store/apps/details?id=com.interpreter.app&hl=ja 01 Light . . . “Continue ( )” “I have backed up all my files ( )” “I understand the safety implications of running AI generated code on my computer (AI )” “I have read and agreed to the 01 App terms of service (01 )” “Tap to scan ( )” 2025 13 01 ∼ — 2025-05 – p.18/21
19. 01 ( ) $ cd 01/software $ poetry run 01 --server livekit --qr --multimodal QR 2025 13 01 ∼ — 2025-05 – p.19/21
20. 01 Light “Tap to scan” ( ) QR QR INFO DEBUG WARNING control+C “poetry run . . . ” / ( ) 2025 13 01 ∼ — 2025-05 – p.20/21
21. 2025 13 01 ∼ — 2025-05 – p.21/21